It seems you have JavaScript disabled.

Ummm.. Yeah... I'm going to have to ask you to turn Javascript back on... Yeah... Thanks.

Interview With Matthew Claassen (CMT) at Trader’s Narrative

I had the opportunity to chat recently with Matthew Claassen (CMT) about his views on the market and what he sees ahead. Matthew’s work caught my attention because he turned bullish early on in 2009 when very few did so. As well, one of the indicators that helped him at that time was the same one we’ve been discussing here for a while: the distance from long term trend.

Of course, there is much more insight and charts so let’s get started:

TN] Hi Matthew, for those who are not familiar with your work and background, can you tell us a little about yourself?

MC] First off; thank you for the opportunity to be a part of your blog. I’ve been managing portfolios and advising professionals since 1986. I hold the professional designation of Chartered Market Technician (CMT) and am a past Director of the Market Technicians Association (MTA).

Previously I was the editor of “The Technical View” a market newsletter which was subsequently bought out by Lowry Research in 2004. I joined the firm and was the Senior Vice President of Lowry Research. Currently, my company’s name is Claassen Research where I provide two services strictly for institutional portfolio managers; weekly commentary in a publication I simply call my Market Update and private advisor retainer. The weekly commentary is $15,000/year. The private consultation/retainer is $50,000/yr. If you’re interested to learn more, you’ll find more details on my site.

TN] The March 2009 bottom took many by surprise, including the very respected firm of Lowry where you worked at before. But you recognized that an important inflection point was about to occur. How did you do realize that and what indicators were you looking that helped you?

MC] In retrospect, the March 2009 low was one of the easier market calls I have made in the twenty five years I’ve been involved in the markets. My decision to tell subscribers to prepare for a rally was a combination of three concepts; two indicators and one study.

In February, 2009 the market was approaching 40% below its 200 day average. That’s more extended than at any time since 1932, and just slightly more than the 1937 decline. That illustrated a powerful fear had taken hold of investors. The contrarian in me felt that regardless of the long term trend, the market was unlikely to sustain such powerful downside momentum. In addition, I had previously completed a study of multi month bear market rallies from 1960 through 2008. There were several interesting concepts that came out of that study, but what was most pertinent was that approximately 40% of all multi month bear market rallies started in March. We all know that the number one bear killer month is October, more bear markets have ended in October than any other month. However, many investors are surprised to hear that March comes in as a relatively close second. Thus, any time the market is declining sharply into the months of October or March, investors should look for signs of a potential bottom.

Lastly, I keep a measure of both Buying Volume and Selling Volume for the broad market, major economic sectors and select international markets. When monitoring Buying and Selling Volume we can typically see a rise in Selling Volume in the weeks ahead of a major market top. The increased Selling Volume is an indication of the investor profit taking into a rally that often precedes a market turn. Conversely, as a market approaches sustainable bottoms we typically see rising Buying Volume, which indicates that prices have fallen far enough to spark investor interest and create demand. In the case of the March, 2009 low; Buying Volume in the US broad market has shown some signs of life going into the November, 2008 low, but really started to pick up steam in early February, even though the market was declining sharply. Investors had started buying into weakness.

Matthew Claassen Inteview Buying Selling Volume chart Mar 2009
(Click to see a larger version in a new tab)

So combining the fact that the market was extremely oversold relative to its 200 day average, that 40% of multi-month bear market rallies start in March and that Buying Volume was increasing all came together as part of the decision. To be fair, I had no idea that the market would rally 70%. My view was that the market would at least rally for a couple months and recover about 38% of the decline. But that meant switching from short to long and letting the market tell us how long the ride would last.

TN] This cyclical bull market (since March 2009) has been one of the most hated ones I’ve seen in my time. Most traders and investors have persistently denigrated it or ignored it as we can see in the lack of mutual fund inflows (and actual outflows!). However, at the same time we have seen some extreme bullish readings in sentiment surveys, especially most recently at the end of the year and start of the new year. How do you reconcile these two conflicting elements?

MC] That’s a great question. I believe that the poor disposition some investors have had toward the market is a reflection of the lack of confidence in the long term outlook, while the bullish readings are a reflection of what the market has accomplished since the March low. I believe sentiment indicators do have value. However, this conflict is a perfect example of why investors should simply focus on the technical aspects of the market that tell us what is happening, rather than what some believe should happen.

A perfect example of this is the persistent low total volume that has plagued this rally for months. We all know that low volume as a market advances is historically bearish, and yet we have experienced one the sharpest gains in market history. If, rather than focusing on total volume we divide volume into Buying Volume and Selling Volume we can see that Buyers have controlled this market from the beginning. I consider Buying Volume the most reliable surrogate measure of investor demand and Selling Volume the best measure of investor supply. Simply put, markets rise when demand is greater than supply. It doesn’t matter if both are falling or both are rising. It is the ratio of demand to supply that defines the trend, not the level. We can see that in the chart below, where both Buying and Selling Volume have been in a down trend since mid-May. But, Buying Volume has persistently outpaced Selling Volume, leading to higher prices. We can also see that the strength of Buying Volume relative to Selling Volume has been weakening since late October, reflecting a tiring trend.

Matthew Claassen Inteview Buying Selling Volume chart Jan 2010

TN] What are your thoughts on the allegation made by Charles Biderman of TrimTabs that the Federal Reserve is manipulating the market by buying futures?

MC] Well, as far as monitoring where the buying has come from, I don’t know of any group that has the resources to match TrimTabs. I am in no position to argue against his findings. However, because his conclusions are based on a process of elimination rather than hard evidence, it will always be controversial.

TN] In your January mid-month market update, you wrote:

The broad market’s intermediate trend remains positive as the major market indices continue to post higher highs and higher lows. While we expect the intermediate trend to eventually run into headway, there is little indication of the increased broad market profit taking that typically precedes an important market top. That said, we do see signs of skittishness within the trading of numerous individual securities. It seems investors are quick to take profits in stocks that have advanced sharply, preferring to sell breakouts and buy pullbacks.

The much awaited correction finally arrived last week. From a short term perspective, how do you see it playing out and what are some of the indicators that you will be using to monitor it? is this it for the aging rally? or do you think it is still ongoing?

MC] Interesting that you say: “The much awaited correction finally arrived last week.” It seems that market sentiment has gone from overly bullish to “This is it” overnight, and that has my contrarian sensors all lit up. The market has only been down for three days.

I have been saying since November 20th that over the intermediate term prices should be lower than they were at the end of 2009 and start of 2010. But in the short term there was very little of the typical signs of profit taking that precede a major market top. In fact, Selling Volume was increasing for only about a week leading into this top. Last week the market suffered its strongest three day decline since before the market rally started in March. The S&P has violated its uptrend line from the June low while holding above and testing a smaller uptrend line from August. The Dow Jones Industrial has closed below its uptrend support. Obviously something has changed and I believe the probabilities are strong that the broad market indices will test their 200 day average in the weeks ahead.

The big question most investors have is “is this the end of the rally from March or just a meaningful interruption?” Unfortunately, in the investment world the only certainties we work with is what has already happened. From an investment management point of view this is a time to hedge the long positions you want to keep and build short positions in the weakest parts of the market. From there we will see how far this market will take us before buyers start to accumulate positions again.

As far as indicators are concerned, I will be watching a few things that might help us determine the strength and potential longevity of this decline. One indicator I will watch closely is Breadth. There are many ways to monitor market Breadth. The Advance Decline Line is one. However, the Advance Decline Line has not been a leading market indicator since before the 2000 top. I prefer to watch an indicator developed by Dr. Marty Zweig many years ago called Breadth Thrust. My own twist is to overlay a 50 day moving average on top his Breadth Thrust indicator.

Matthew Claassen Inteview NYSE OCO breadth thrust chart
(Click to see a larger version in a new tab)

As a side note, Breadth Thrust is simply a 10 day average of Advances / (Advances + Declines). This is from an old introduction to the Breadth thrust. I am not sure when it was written, but I know it was in the 1990’s:

According to Dr. Zweig a Breadth Thrust occurs when, during a 10-day period, the Breadth Thrust indicator rises from below 40 percent to above 61.5 percent. A “Thrust” indicates that the stock market has rapidly changed from an oversold condition to one of strength, but has not yet become overbought.

Dr. Zweig also points out that there have only been 14 Breadth Thrusts since 1945. The average gain following these 14 Thrusts was 24.6 percent in an average time frame of 11 months. Dr. Zweig also points out that most bull markets begin with a Breadth Thrust.

If the current weakness is the beginning of a decline that will be more than just to test the 200 day average or a 38% retracement of recent gains, then I would expect to see a stronger negative reaction than how the indicator reacted in October, and the follow through should be consistently weak enough to bring the 50 day average below 50. In other words, strong selling is not followed by very strong buying. This would be similar to how the indicator responded at the top in 2007 and in 2000.

TN] We’ve been discussing the ratio of new highs as a yard stick to help us to find major market tops. Is this something that you watch? if so, how do you interpret it right now?

MC] I do monitor New 52 Week Highs and Lows, but perhaps differently than most. Like many investors with an economics background, I like to divide my indicators up into three categories; leading indicators, coincident indicators and lagging indicator. Each has an important role to play in determining a change in trend and the sustainability of the trend. I view New 52 Week Highs and Lows as a lagging indicator that confirms a trend. If your leading and coincident indicators have fallen in line with the markets trend change, when a lagging indicator like New Highs confirm an uptrend or increased New Lows confirm a downtrend, you can have confidence the trend is sustainable

TN] Taking a broader look, where are we in terms of market cycles? is this a real bull market? or are we still within a long term secular bear market that isn’t quite finished?

MC] Back in early 2000, when I was a portfolio manager, I wrote an article for current clients that defined my views on secular bull and bear markets because at that time I believed the secular bull market was about to end. Historically, the US equity markets have changed secular trends about every 18 years.

Matthew Claassen Inteview 18 year cycle chart
(Click to see a larger version in a new tab)

Then, in 2002 I wrote a piece that I called “The Three Bears” that detailed the differences between three different types of bear markets (Inflationary Secular Bear, Deflationary Secular Bear and Cyclical Bear). The approximate 18 year period is important because historically the investment losses in secular bear markets have been significant enough that it takes an entire new generation of investors that did not experience those losses to step in and generate the demand needed for a new secular bull market.

In brief, I believe a secular bull market is a period of time where the earnings trend and the trailing P/E ratio trend are expanding. Historically, the price gains in secular bull markets are due more to an expanding P/E ratio than increased earnings. I also consider the trailing P/E ratio to be a long term sentiment indicator.

One need only look at the trend in earnings growth since 2007 and we can tell this is not a secular bull market. Also, historically new secular bull markets start with a trailing P/E ratio below 10 and rising earnings. In other words, market sentiment was declining enough to drive the P/E ratio to an extreme low despite improved earnings. At the end of secular bear markets it was the rising earnings plus a weak market that brought the P/E ratio to below 10. The current market has not even come close to approaching these historical norms.

This means the US equity market is still in a secular bear trend and how it behaves from here will depend a lot on whether there are underlying inflationary or deflationary pressures. Past inflationary bear markets have seen a declining real rate of return of near 80%, but maximum nominal declines closer to 50%. Two periods where this occurred was the 1906-1921 inflationary secular bear and the 1968 -1982 inflationary secular bear market. Deflationary bear markets tend to also have 80% real loss coupled with an 80% nominal price decline. That was the case in the US market from 1929 - 1932 and in the Japanese market from 1990 – 2003.

At this point, all I can say about the US secular bear market is that it appears to be a combination of both past inflationary and deflationary bear markets. As such, I expect we are in uncharted territory.

TN] Thank you very much for your time Matthew, that was very enlightening.

MC] Thank you.

Enjoyed this? Don't miss the next one, grab the feed  or 

                               subscribe through email:  

29 Responses to “Interview With Matthew Claassen (CMT)”  

  1. 1 Samuel

    Very interesting. Thanks for doing this interview.

  2. 2 freicks

    Uncharted Territory indeed. You can never know for sure we we are going but I find Dr. Martin Zweigs indicator useful and will certainly be taking a look at that. Thanks so much for a great interview with TN.

  3. 3 Ernest Harris

    Yes, Thank you very much for doing and publishing this interview!

  4. 4 Avi

    great interview… any idea on how to calculate ‘buying volume” vs “selling volume” .. i never really understood this.. in every transaction there is both a buyer and a seller.. I guess it more like who is hitting the bid/offer

    still i cant find buying volume on any of my esginal charts

  5. 5 Ross Edwards

    I am not sure if this is regarded as “Buying or Selling Volume” but it is something I use and call “Cashflow Stock”.

    I use it in my charting program, AmiBroker. No doubt the other charting programs have variations on this theme.

    It does what I want it to do.

    /* CASH FLOW DIRECTION. Calculates the dollar inflows/outflows of advancing & declining stocks */

    A1=IIf(C>Ref(C,-1),(V*C),0) ;

  6. 6 Ross Edwards

    This is the rest of the above :
    A1=IIf(C>Ref(C,-1),(V*C),0) ;

  7. 7 van

    Excellent interview, thanks.

  8. 8 wayne

    I spent some time toying around with the Zweig index mentioned in your post by Classen.

    My study was based on the available tape data that I have from Jan 01, 1970 through Jan 31, 2010.

    The data includes the non stock related issues posted daily by NYSE, which most major research firms now omit.

    For the data that I had available, the most impressive combination of variables that I found was

    1. When Index was greater than 64.0

    2. Coming from a reading below 42.0 in the last 14 days.

    There were 12 signals in the last 40 years.

    03 months later, the S&P was 10-2 for an avg gain of 08.04%
    06 months later, the S&P was 12-0 for an avg gain of 17.11%
    12 months later, the S&P was 12-0 for an avg gain of 22.22%
    18 months later, the S&P was 12-0 for an avg gain of 24.74%
    24 months later, the S&P was 10-2 for an avg gain of 27.48%


    1. The signals were perfect in predicting up markets 6, 12, and 18 months in advance (one year equals approximately 252 trading days).

    2. We had looked at other thrust studies in the past. This one is different in that it requires that the thrust be launched from oversold condition.

    3. Since we are now some 320 days from the last signal on Mar 23, 2009, the relevance of this particular thrust study to today’s market is questionable, although very interesting.

  9. 9 wayne

    One correction,

    we are now some trading days 220-230 days from the last signal

  10. 10 Ernest Harris


    What software did you use for your analysis of the Zweig Breath Thrust Index?

  11. 11 wayne

    I am a retired engineer that writes my own software old school via a fortran compiler. This gives me incredible flexibility. For example, for this study, I wrote a 100 line code to

    1. Calculate the daily a/(a d)
    2. Told the computer to vary the upper zweig threshhold from 55-65,
    3. Vary the lower threshhold from 35 to 45
    4. Vary the number of days to look backwards for the trailing lower threshhold from 10 to 20
    5. Print out any of the 1000’s of combinations that had a perfect record 252 days ahead.

    I have a degree in something called Operations Research which has a lot of optimization theory and prob & stats. I had no idea at the time of acquiring my education, that it’s best application would eventually be studying the stock market. If I were a bit younger, would have loved to gone to work for one of these research firms with tons of available data, but I am able to keep myself entertained with the databases I have been able to put together. For a younster, interested in doing this for a living, I would highly recommend a O.R. Degree with a minor in Economics or vice versa. Then I would have been dangerous.

  12. 12 Ernest Harris


    Thanks very much for you reply.

    I too am retired. I was an applications analyst/programmer. I mostly coded in COBOL.

  13. 13 wayne

    I didn’t submit the table of results for the above study because the space formatting usually doesn’t come through readable in the comment section, but I will attempt below. Should read, the Date, Zweig index modified, and then 5 columns of results for 63,126,252,382 and 504 days, chosen to approximate to 3, 6, 9, 12, 18 and 24 mts. (252 trading days is approximately 1 year, give or take a day)

    DATE A/A D 63DYS 126DYS 252DYS 382DYS 504DYS
    700828 65.26 04.97 18.49 22.75 29.71 35.62
    711208 66.41 12.42 11.07 21.76 10.09 -0.42
    741014 65.62 -1.46 18.64 22.99 37.75 39.73
    750107 64.39 14.04 31.50 31.70 45.79 48.83
    760106 67.26 10.67 10.70 13.01 07.01 -0.01
    790112 64.18 02.07 02.39 10.00 17.92 33.61
    800411 64.06 13.54 26.25 29.60 17.02 11.76
    820823 66.93 18.03 27.49 41.25 34.15 41.39
    840807 64.37 03.60 10.99 15.49 30.88 45.55
    910130 64.36 11.55 13.76 20.37 23.85 28.51
    040601 64.86 -1.52 04.69 07.22 12.15 14.67
    090323 68.88 08.52 29.38 30.50* 30.50* 30.50*

    As I mentioned in above comment, since we are now some 220 days from the last signal on Mar 23, 2009, the relevance of this particular thrust study to today’s market is questionable, although very interesting.

    Up-Down 10-2 12-0 12-0 12-0 10-2
    Average 08.04 17.11 22.22 24.74 27.48

    * Still in Progress

    I’ll look for a sell signal soon.

  14. 14 Ernest Harris


    Thanks very much. I was able to pastes the results into Microsoft Word and convert to table. It reads very nicely!

  15. 15 DaveinPhilly

    I would think that this breadth thrust idea also ties into Lowry’s concept of 90% upside days, where 2 or more upside days close to each other is very bullish. Lowry’s won the MTA award for this research and you can probably find it on the web.

  16. 16 DaveinPhilly

    In Marty Zweig’s book, “Winning on Wall Street”, on page 85 he talks about his A/D indicator, where he looks at the A/D ratio over a 10 day period. When the ratio was 2/1 or more, there were only 10 instances from 1953 to 1990 (when book was written). 6 month return after such a signal was 15.6%

  17. 17 wayne

    Dave in Philly

    See my post in Sept, on the ‘Mother of Thrust Years’ re a followup to Zweig’s thrust work.

    I have researched the 9 to 1 days as well, but I havent had a lot of luck with them in the last decade. If you read Desmonds (Lowry) MTA paper on it, there are some caveats as to how it is to be interpreted. Maybe I simply havent studied it properly.

    I have had some success with a 5 day moving average of up volume to total volume.

    Zweig was the first very major influence in my studies, but with all the respect in the world I have for him, his monetary models have failed terribly in the last decade. At least the ones outlined in his book. I suspect due to the fact that they were backtested over a high interest period. He has made tons of money and the best I can tell is laying low now.

  18. 18 wayne

    Reverse Zweig Thrust Signals.

    I took the logic and reversed it to see if I could get a good sell signal system and I couldn’t come up with anything better than 50/50. This is likely due to the fact that

    1. Tops are not normally established in the same fashion as bottoms.

    2. The stock market over any 30 year period has a bullish bias, thus making reliable short signals more difficult to identify.

    Unless someone has some insight ow, it is my opinion that using the Zweig index for sell signals would require using a different approach or at least something different than the 10 day moving avg. I may revisit, but for now, on to other endeavors.

    Now this is interesting, just before I hit send, it occurred to me to test whether since the reverse thrust signals did not give good sell signals, what kind of buy signals did they give???

    Zweig index below 44 after being at 62 in the last 10 days is an solid buy signal with a 6-0 track record. This makes a lot of sense, buy pullbacks after a positive thrust signal.
    S&P % S&P %
    Date Index 252dys 504days
    741028 42.70 28.02 42.77
    811026 42.89 12.83 40.45
    821123 42.79 24.92 22.69
    860501 42.33 22.62 11.67
    090120 41.33 62.33 35.07
    091002 42.90 06.08* 06.08*

    * in progress

    Now, of course this is only 6 data points, but we may have stumbled onto something because this actually has some relevance to today’s market since we got a 62 reading in late sept 09 and then the 42.9 on Oct 02 and per our study should be up solid double digit a year later on Oct 02, 2009. This would support a lot of our other work as well.

    Reliable sell signal systems are priceless.

  19. 19 DaveinPhilly

    “”Zweig was the first very major influence in my studies, but with all the respect in the world I have for him, his monetary models have failed terribly in the last decade.”"

    For many years, I used to trade off the commitment of traders report, the commercial hedger category for the S&P 500. It used to work very well and I made a bunch of money. It stopped working in the fall of 2003. Not sure why, probably the advent of ETF’s.

  20. 20 DaveinPhilly

    wayne, you said..”Now, of course this is only 6 data points, but we may have stumbled onto something because this actually has some relevance to today’s market since we got a 62 reading in late sept 09 and then the 42.9 on Oct 02 and per our study should be up solid double digit a year later on Oct 02, 2009. This would support a lot of our other work as well.”"

    Nice work! But I think your date is wrong. Don’t you mean Oct 02, 2010?

  21. 21 wayne

    If trading could be turned into a science, Einstein would have been a billionaire.

  22. 22 MatthewClaassen

    There are a lot of great questions and thoughts on the interview. I am very pleased to see so many educated and interested readers.

    I want to try to answer some of the questions posed above. If I miss any, please let me know.

    When I calculate Buying & Selling Volume for an index, I actually calculate the data for each of its components. The aggregate of the components is what we use to determine true buying and selling. This is a very different approach than just looking at the volume of an index or ETF and the results are much more useful.

    I have heard that a number of readers were interested in where they could find an NYSE Operating Company Only Index. As you know, over 50% of the current NYSE Composite is comprised of ADR’s, preferred stock, ETFs etc. I have found that if you simply use the S&P 500, or better yet the S&P 1500 as a substitute the resulting data for advance decline lines and other breadth studies will be a near perfect correlation. I have rune breadth on both for more than 10 years with 40 years of back history. I believe the NYSE OCO and the S&P 500 data are interchangeable. The key is to use a broad index that has a healthy mix of the ten major economic sectors (i.e. Materials, Consumer Discretion, Energy etc.)

    Lastly, I intended to show that the Zweig Breadth indicator is a useful measure of market health beyond the standard Breadth Thrust signal. I overlaid a 50 day average to show that the Breadth Ratio tends to deteriorate before market tops and strengthen before market bottoms. It can be a nice tool to add to the host of volume, price and volatility indicators already at your disposal.

    Best Regards,

    Matthew Claassen, CMT

  23. 23 MatthewClaassen

    One last comment regarding Wayne’s post above:

    Actually markets do not top and bottom the same way. That is why the Zweig breadth thrust methodology works on bottoms and not tops.

    Market bottoms are more of an “Event”. They are filled with sharp emotion and, relative to tops, happen very quickly.

    Market Tops are a process. Typically they are a slow erosion of stock momentum. In fact, by the time most major indexes have posted a bull market high, more than 20% of individual issues are already down 20% or more from their high. So, the best indicators for identifying most bull market highs are ones that measure breadth on a long term scale. For example, the percent of issues above their 150 day average is a good tool to use as it often diverges prior to a major market high. That said, highs from bear market rallies are different than highs from a four year cycle bull market. We don’t know if the most recent high will lead the market to a complete test of the March lows or not. The fact that there were no significant divergences in breadth suggests that the market is not ready to roll over into a bear yet. But, the rally from March is more charectoristic of a bear market rally than a new bull market. I believe its best to just expect a 38% retracement or a test of the 200 day average. Then look at the indicators to see what’ next.

  24. 24 wayne

    Mr. Claassen,

    Thanks for taking the time to share your insight. I shared with Babak that it was my opinion, that your interview was one of the top 5 he has posted in the last year. If you are not careful, you will get pulled into ongoing debates in this forum on various technical subjects. You have been warned, so I will give it a shot.

    Since you referenced Dr. Zweig’s thrust index and are obviously familiar with his traditional breadth thrust research, I would be curious how you are able to reconcile your bearish outlook with the three traditional Zweig Momentum Thrust signals that were observed in 2009 (March 23rd, July 23 and as late as Sept 16). As I’m sure you are aware, single signals have strong statistical bullish significance and double signals (1962, 75, 88 and 09?) even more so.

    Any comment?

    Traditional Zweig Thrust signal defined as a ten day period where advances lead decliners by a 2 to 1 margin on the NYSE

  25. 25 MatthewClaassen


    Thanks for the warning!

    Regarding my views and the Zweig Breadth Thrust Indicator:

    First my views should be separated into time frames:

    Long Term: I believe there is little doubt the US is in a secular bear market. Because past secular bear markets have not ended until after 1.The trailing P/E is under 10 and 2. Earnings are rising – I believe we still have some ways to go.

    Near Term: I believe the probabilities are that the market is in the process of testing its 200 day average or an overshoot to a 38% retrace of the gains from March. More than that? As I said in the interview; we’ll see what kind of buying steps in once the markets have corrected closer to the 200-day average.

    Studies of the Breadth Thrust Indicator show the market up for more than a year from the “thrust”. The traditional indicator is defined as “According to Zweig, a “Breadth Thrust” occurs when the Breadth Thrust indicator rises from below 40% to above 61.5% within 10 trading days.” It was created as a method to identify New bull markets, thus its best studied use is when it occurs after a major decline. This is different than what you noted on your comment. I do not consider a Breadth Thrust to be defined as a ten day period where advances lead decliners by a 2 to 1 margin on the NYSE. I don’t recall having heard that definition before.

    Still, a test of the 200-day average and further rally would fit with that bullish scenario. My opinion does not contrast with that. Rather, my experience forces me to be aware that the market does do things it has never done before. For example; did you know that the US equity market had NEVER made a final bull market high in the month of October until 2007? Thankfully, I was putting more emphasis on my indicators than the calendar at that time. When we get our mind stuck with the concept that the market can’t do something because it never has before, be open ourselves up to big surprises. Big surprises mean big losses. I prefer to let the market tell me what it is doing, while using seasonal tendencies and past studies as a guide, not a rule book.

    In another post Baback had commented on an NDR study showing that markets have never made a final bull market high coincident with record New Highs. (Actually there have been a number of research houses that have done this study over the years, all with the same result.) Because January had record New Highs, it would be unusual for the January high to be the final high of this cyclical bull market. He also indicated that bearishness picked up very fast after only a few down days. These are both excellent observations. While these stats may be consistent with a pullback to the 200day average, they also reduce the probability that the market has made a final high. But they do not eliminate that possibility.

    Lastly, a study I did recently found that the second best time to buy stocks is at a successful test of a 200 day average. (the best being a bear market low) The study looked at six month forward returns and allowed for an index to slightly overshoot the 200 day average in a correction. So, if the market declines to the 200 day average or 38% retracement level and my indicators show accumulation, I will be aggressively long. I have laid out all the above to my subscribers’ so they have the ability to make the decisions they need to make. If, instead, I insisted the market will do X or Y, because that is what it normally does under these circumstances, I would be doing them a disservice and leave them ill prepared to for those times when the least probable outcome becomes reality.

  26. 26 wayne

    Mr. Claassen,

    Again let me thank you for myself and all the readers of this blog for sharing your research and comments to us at a tremendous discount to those who normally subscribe for such insight. I had been giving some thought to converting my breadth analysis research to S&P 500 data and your suggestion to do such has given me reason to move it up on my todo list.

    It was presumptious of me to assume the 2/1 Ten Day Breadth Thrust was the ‘traditional’ Breadth Thrust indicator in technical circles simply because it was the one I had been most familiar with. I originally viewed the 61.5/40 version as a subset of the one that I was more familiar with because it required that the launch be initiated from a somewhat oversold condition (index

  27. 27 DaveinPhilly

    Matthew said..”When we get our mind stuck with the concept that the market can’t do something because it never has before, be open ourselves up to big surprises. Big surprises mean big losses. I prefer to let the market tell me what it is doing, while using seasonal tendencies and past studies as a guide, not a rule book.”"

    Thank you! Let’s not forget that the overriding fact of all these historical backtesting of indicators is that the sample size is woefully too small. Forget about the black swans, with these small amount of instances (of sample size), we haven’t even seen the pink swans yet. One hundred years of stock market data is certainly not enough when talking about multi-year cycles such as bull and bear markets.

  28. 28 wayne

    cont from above ( it got cut off for some reason)

    I googled Zweig Thrust indicator and saw where the 62/40 version referenced prominently. So I learned something this week. Thanks again.

    My introduction to Breadth thrust came in the mid 80’s in Dr. Zweigs book ‘Winning on Wall Street’ in the section on momentum, page 85 in my version. It is likely that he published the more simplistic 2/1 version in his book because it was easier to understand for the masses he was attempting to sell it too. I have seen Dan Sullivan of the chartist reference it prominently over the last year. NDR now uses a derivative of it on a index they derived that includes Nasdaq issues.

    The 2/1 index most likely lost favor with analyst because there were no sightings between 1991 and 2009. When they started to occur in 2009 again, I posted two blogs here, documenting their history. ‘History of Momentum Thrust’ and the ‘Mother of all Thrust Years’.

    I personally have discovered that a five day A/(A D) over 73.5 is 12-0 for twelve month forecast, using nasty NYSE data since 1970 has an incredibly strong recent record. A 13th signal on Sept 10th, 2009, needs the market to hold onto current gains between now and Sept 2010 to maintain it’s perfect record.

    Lets see how it plays out.

  29. 29 wayne

    yes , I meant 2010

Leave a Reply